Spatiotemporal dynamics of striatal dopamine release in reinforcement learning

描述（由申请人提供）：经典条件反射可以说是我们拥有的使我们的行为适应不断变化的环境的最基本的工具。食欲经典条件反射可以被认为是在强化和环境中表明其可用性的刺激之间建立学习关联的过程。这种形式的“强化学习”为生物体提供了一种有效的方法，可以使用预测性信息来塑造和指导准备行动，以最大限度地提高回报。在过去的三十年里，我们对强化学习背后的神经生物学机制的理解取得了重大进展。特别是，多巴胺神经元活动的短暂“阶段性”爆发，被认为会在纹状体中产生细胞外多巴胺浓度的快速和短暂增加，作为自然和药物奖励强化过程的必要组成部分，已经受到了相当大的关注。儿茶酚胺研究的中心教条认为，多巴胺的传递是作为一个统一的广播信号从中脑传递到前脑的所有目标结构。然而，越来越多的证据表明，纹状体内的不同核可能在学习过程的不同阶段对主要奖励和条件提示的反应中接收到不同的信号。据推测，多巴胺信号的动态和稳定性的这种区域特异性与多巴胺在这些结构中可能发挥的学习和行为控制中的主要分离作用相对应。目前的提议将以三个具体目标来检验这些假设。目的1将研究学习历史和任务的具体特征对腹侧纹状体中阶段性多巴胺稳定性的影响。目的2将研究在学习的不同阶段，阶段性多巴胺对纹状体激活的影响，并评估行为的一致区域控制。目的3将在强化学习的多个阶段（习得、扩展训练和消退）中检查腹侧和背侧纹状体的阶段性多巴胺释放，目的是将特定行为的发展与每个结构中的阶段性信号的特征相关联并进行比较。

DESCRIPTION (provided by applicant): Classical conditioning is arguably the most fundamental tool we possess for adapting our behavior to a changing environment. Appetitive classical conditioning can be thought of as the process that establishes a learned association between reinforcement and the stimuli in the environment that signal its availability. This form of "reinforcement learning" provides organisms with a potent means to use predictive information to mold and guide preparatory action in an effort to maximize reward. The past three decades have been marked by major advances in our understanding of the neurobiological mechanisms that underlie reinforcement learning. In particular, short "phasic" bursts of neuronal activity in dopamine neurons, which are thought to produce rapid and transient increases in extracellular dopamine concentration throughout the striatum, have received considerable attention as a necessary component of the reinforcement process for both natural and drug reward. The central dogma in catecholamine research has held that dopamine transmission proceeds as a uniform broadcast signal from the midbrain to all target structures in the forebrain. However, there is mounting evidence that separate nuclei within the striatum may receive differential signals in response to primary rewards and conditioned cues at different stages in the learning process. It is hypothesized that this regional specificity in the dynamics and stability of dopamine signaling corresponds to the largely segregated roles in both learning and behavioral control that dopamine may play in these structures. The current proposal will test these hypotheses with three specific aims. Aim 1 will examine the impact of learning history and specific features of the task on the stability of phasic dopamine in the ventral striatum. Aim 2 will examine the influence of phasic dopamine on the activation of the striatum during different stages of learning and assess the coincident regional control of behavior. Aim 3 will examine phasic dopamine release in the ventral and dorsal striatum during multiple stages of reinforcement learning (acquisition, extended training and extinction) with the goal of correlating and comparing the development of specific behaviors to the profile of phasic signaling in each structure. PUBLIC HEALTH RELEVANCE: Although adaptive under normal circumstances, reinforcement learning is one of several contributing factors to the powerful control drug-related stimuli have over drug-seeking and drug-taking behaviors in many models of addiction. The phasic activation of the neuromodulator dopamine is implicated in both reinforcement learning and many aspects of drug abuse. The primary goal of this proposal is to examine the contribution of phasic dopamine release in specific brain regions to reinforcement and behavioral control.